Soil compaction induced by vehicle traffic has aroused more concerns due to its negative impacts on soil functions and ecosystems. Replacing tires by tracks is considered as a technical method to mitigate the soil compaction, which increases the vehicle's contact area and decreases the mean ground pressure. However, the interactions between the track and the soil are complex, the stress distribution at the interface between the track and the soil is uneven, which may reduce the effectiveness of the track in decreasing soil stress. Thus, to determine the ability of track to reduce the soil compaction compared with tire is importance. The objective of this study is to investigate the impact of undercarriages (tire vs. track) on the magnitude of soil stress and soil functions (i.e. pre-compression stress, air permeability, dry bulk density), as well as the impact of vehicle velocity on the magnitude of soil stress. The test was conducted on the sandy loam soil. Three repeated measurements were carried out on tracked harvester and tractor with similar axle load. To measure soil stress, the load transducers were embedded in the centerline of the tire and track at a depth of 0.15 and 0.35 m respectively. After vehicles pass, soil samples are collected at the depth of 0.15 and 0.35 m in the test area. Soil pre-compression stress, air permeability and dry bulk density were measured in the lab. The soil stress in the depth range of 0.1-0.7 m was calculated by using the soil compaction model. The results show that compared with tires, the vertical and horizontal soil stresses are reduced under the track compaction. However, the reduction of the vertical stress was greater than that of the horizontal stress. Under the effect of tyre compaction, the average vertical stress at the depth of 0.15 and 0.35 m is 2.2 and 2.0 times of track compaction respectively, whereas the average horizontal stress of tyre compaction was only approximate 1.2 and 1.1 times larger than that of the track respectively. Interestingly, no difference in vertical and horizontal stress was found between the track and the tire compaction at the depth of 0.7 and 0.4 m respectively. It indicates that the track is more effective in reducing the soil compaction for the topsoil than for the subsoil. The vertical and horizontal compaction stresses of tire and tracked vehicle decrease with the increase of vehicle speed, but the speed of stress reduction under the track compaction is faster than that of tire. The measured air permeability of track compaction is obviously larger than that of tire, whereas the measured dry density and pre-compression stress displayed no significant difference between the tire and track compation. In other words, the ability of the track to reduce soil compaction is weakened due to the uneven stress distribution and the longer compaction time. Though the calculated soil stress is generally in line with the measurement results, the measured vertical stress is lower than the calculated value for the tire compaction but higher for the track compaction, the measured horizontal stress are lower than the calculated values for both the tire and the track compaction. The accurate prediction of the distribution of stress on tire/track surface and the reasonable selection of concentration factor are the key to model calculations. Future research may focus on the impact of support roller configuration and diameter on the distribution of soil stress of track compaction. The research results are helpful to improve the uniformity of soil stress distribution under the track compaction, so as to reduce soil compaction.